Why do we over-eat?
The answer likely lies not in our bellies, but in our brains.
Now a team Kiwi scientists have begun a new study that aims to pinpoint a "sensory fingerprint" behind the urge to eat for pleasure.
New Zealand has the third-highest rate of obesity among OECD nations, with almost one in three adult Kiwis considered obese, and one in 10 children.
It wasn't always this way, said Dr Mei Peng, of Otago University's Department of Food Science.
"Have you ever looked at old photos, and wondered why people had such uniform physiques before the mid-1900s?
"Over recent decades, human body sizes have diversified enormously."
In a modern environment where food was becoming ever more accessible and palatable, fighting against our desire to over-indulge could be a very challenging task.
Intriguingly, some people were particularly susceptible to eating for pleasure, or what's termed hedonic eating.
While these differences were thought to be related to brain networks responding to reward, it was unclear why food holds greater rewards for some people than for others.
"Eating is a multi-sensory experience, where the taste, smell, appearance and even sound of food are integrated to give pleasure," Peng said.
"Our tantalising new findings suggest that we each have a unique sensory fingerprint that portrays a person's relative sensitivities across senses."
Peng and her colleagues predict that this fingerprint controls reward-related brain networks and determines individual susceptibility to hedonic eating.
In a new $300,000 study, supported with a Marsden Fund grant, the team will attempt to find the neural basis of food-related problems New Zealand is now grappling with.
The project would combine experimental psychology, food sensory research and neuro-imaging techniques to test the relationships between sensory fingerprints and hedonic eating behaviour, using a large sample of study participants.
"Our project relies on a very novel technology set-up which enables real-time brain scans during exposure to multi-sensory stimuli, such as smells, tastes and texture," she said.
"We are currently at the testing phase of this challenging set-up, and are excited to apply these state-of-the-art tools.
"In addition to the practical benefits of developing a sensory 'fingerprint', the project will generate novel knowledge about the interaction between the senses and human eating behaviour, which may lead to new targeted intervention to help with obesity."
The new study comes after leading health academics called for a major improvement in food policy to tackle obesity, which was projected to have already overtaken tobacco as the leading risk factor causing health loss.
The second Auckland University Health Food Environment Policy Index, published in 2017, saw the country score poorly on about half of the food policy indicators used to measure the policy changes needed to tackle the problem.
The report asked a panel of 71 independent and government public health experts to rate the extent of implementation of Government policies on food environments and infrastructure against international best practice.
Forty-seven per cent of all the good practice indicators were rated as having "low" or "very little, if any" implementation.
Labour has previously advocated for "real targets" for obesity reduction, rather than referral targets, but it remained unclear what those targets, if set, would look like.